Central Bank decision-making process: the Bank of England and the ECB - National Institute of ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Central Bank decision-making process: the Bank of England
and the ECB
9 March 2022
Maria Demertzis* Catarina Martins* Nicola Viegi**
* Bruegel ** University of Pretoria
The way central banks decide on monetary policy has evolved over the years. One important shift has
been assigning monetary policy decisions to a committee rather than to one individual. In May 1997,
the Chancellor of the Exchequer, Gordon Brown, granted the Bank of England operational
independence, in other words, the independence to decide how to achieve price stability. By contrast,
the Bank would not be granted target independence as the definition of price stability, the inflation
target, would remain with the government. The Bank of England Act1, which came into force on June
1st, 1998, dictates that the monetary policy measures to achieve the inflation target are to be decided
by the Monetary Policy Committee (MPC). The MPC would comprise of the Governor, two Deputy
Governors, two members appointed by banks, and four external members appointed by the
Chancellor.
What makes for an effective decision-making process?
Table 1 below summarises the decision-making process of three central banks: the Bank of England
(BoE), the European Central Bank (ECB) and the Federal Reserve Board (Fed). The pursuit of price
stability by a committee is by now the standard in Central Banking.
The move towards committees coincided with the shift to central bank independence (Blinder, 2007).
This was the natural consequence as Central Banks no longer took orders from their governments.
Instead, they needed to pool information that would help them make better decisions in uncertain
circumstances - a necessary step when performing complex tasks, like monetary policy.
But, as Blinder indicates, the decision-making by committees is not identical in all central banks.
Broadly speaking, they fall into two categories: individualistic, where every member is asked to vote
and those votes are then revealed; and collegial, where all members come into one decision without
attributing votes. The Bank of England and the Fed System are both prime examples of the former. An
individualistic system is built on the diversity of views and the methods on which such views are built
reduce the risk of group thinking. On the other hand, however, when votes are split, they face the
challenge of communicating to the public the rationale behind any decision effectively. The ECB, by
contrast, is a collegial-based system, where the decision reached is presented as that of the whole
decision body. The emphasis is on communicating one view and therefore claiming ownership by all
who participate.
1https://www.bankofengland.co.uk/-/media/boe/files/quarterly-bulletin/1998/the-boe-
act.pdf?la=en&hash=245383B3380F691849B043BE87D31CD2E0EA1256
1Table 1: The decision-making process of three main Central Banks
BoE ECB Fed
Decision-making
Monetary Policy Federal Open Market
Decision-making body Governing Council
Committee (MPC) Committee (FOMC)
Number of members 9 25 12
7 members of the Board
6 members of the of Governors + the
The Governor + 3 Executive Board + the president of the Federal
Composition of the Deputy Governors + the governors of the Reserve Bank of New
decision-making body Chief Economist + 4 national central banks of York + 4 of the
external members the 19 euro area remaining 11 Reserve
countries Bank presidents (1-year
term on a rotating basis)
7 male + 2 female (both 23 male + 2 female (both
Gender distribution 4 male + 4 female**
external members) in Exec Board)
All euro area
Nationalities Mixed but mostly British US citizens
nationalities
6 Exec Board members
vote permanently + The 12 members of the
Who has the right to All 9 members of the
15 votes from 19 NCB FOMC – yearly rotation
vote MPC
Governors – monthly for the RB presidents***
rotating basis*
Reaching a decision Voting Collegial Voting
Disclosure of voting Yes No Yes
every six weeks
Frequency of meetings 8 times a year typically, 8 times a year
(~ 8 times a year)
Notes: The composition described is for 2022, not historical. *ECB rotating voting: since 2015, there are 15 votes gathered
from the 19 National Central Bank Governors on a monthly rotating basis. There are 2 groups of countries: big (4 votes) and
small (11 votes). All (6+19) participate in the discussions. **In view of the recent resignations and no replacement up to this
point, the numbers do not add up to the expected 12 FOMC members. ***Note that all 12 Reserve Bank presidents attend
FOMC meetings and participate in the discussions.
Many in the literature have attempted to identify what makes for an effective committee. Having clear
objectives, efficient instruments and being independent are of prime importance (Maier, 2010). But
having a manageable size, between 5-9 members, is also viewed as preferable (Berger, Nitsch and
Lybek, 2008; Hansen, McMahon, and Velasco Rivera, 2014). In these cases, having a rotation system
may help combine a manageable size with bringing in more information. The diversity of members’
backgrounds, for example in the form of internal and external members, adds to the information set
and can help avoid extreme ideas or indeed group thinking. Last, a voting system that attributes and
evaluates members’ contributions adds to the effectiveness of decisions.
2Linked to these two different types of committee decision-making is also the role of statements and
minutes as part of the communication process. Individualistic committees use the minutes a lot more
as a way of communicating both the decision reached as well as possible disagreements. Detailed
statements, on the other hand, are a more effective tool in collegial central banks. This is because, as
Paul Tucker, a former BoE MPC member, puts it “it is more difficult for us than for some of our peers
to release an informative statement immediately after the policy meeting: if you don’t know what
you’re going to decide, it is pretty hard to prepare a draft in advance!” (Tucker, 2008).
A number of studies then examine how the characteristics of members of monetary policy committees
play a role in decision-making. The literature that investigates the FOMC look at educational and
career characteristics as well as differences in behaviour between political appointments (like the
Governor who is appointed by the US president) and Bank Presidents. Eichler, Lähner and Noth (2018)
find that FOMC members who have a financial industry background or represent a region with a large
banking sector are more sensitive to local banking instability. Smales and Apergis (2016) find, on the
one hand, that the FOMC’s Chair tenure and the experience in Government lead to more dovish
decisions. By contrast, the longer the time working as bank staff, the greater the preference for
hawkish decisions. And indeed internal disagreements are very much attributed to the background
characteristics of FOMC members as well as to political influence (Bennani, Kranz and Neuenkirch,
2018).
Authors that study the workings of the Bank of England’s MPC also analyse the relevance and
importance of dissent. Given the set-up of the MPC, many have studied how the distinction between
internal and external members plays out. Harris, Levine and Spencer (2011) show that external
members are more likely to dissent when there are deviations of the MPC’s inflation forecast from the
target. Gerlach-Kristen (2009) argue that insiders typically attach greater weight to inflation
stabilization than external members, who are more dovish. They attribute this to fact that externals
are appointed (and potentially re-appointed) by the Governor, which gives them an incentive to be
more “recession averse”. Harris and Spencer (2009) show that insiders tend to vote as a block and are
typically on the winning side of policy decisions, provided their numeric superiority. Hansen,
McMahon and Velasco Rivera (2014) also find that internal MPC members have superior expertise
when compared with externals which casts some doubt on the value-added of external members.
However, they and others, particularly Downward and Mearman (2007) caution such a statement as
they point to the importance of triangulation, in other words, the use of diverse sources of information
to inform decisions. Such diverse sources could range from different methods and data to different
theories and investigators.
Last, other streams of literature have focused on different aspects such as the relevance of nationality,
particularly in the context of the ECB, and gender. Badinger and Nitsch (2011) study the ECB and show
that indeed beyond a certain management level, nationality does affect the formulation of monetary
policy. The issue of gender has been subject to growing attention and there is an increasing number
of studies trying to examine to which extent gender affects monetary policy decisions. Rieder (2021)
shows that there is mixed evidence in the literature and warns for caution given the current small
proportion of women in the samples used to investigate this topic.
Individualistic vs collegial: are they really that different?
While many in the literature have attempted to understand what makes for an efficient decision-
making set-up, much less is known and understood as to why Central banks might opt for one system
or the other.
3The answer often lies in reasons that have probably little to do with our knowledge of optimal design
and more to do with culture or broader political economy reasons. Malmendier, Nagel and Yan (2021)
show that, with reference to the FOMC, differences between FOMC members’ inflation expectations
and Fed staff forecasts can be explained by personal inflation lifetime experiences and do affect voting
outcomes.
Arguably, this link between preferences and experiences is the one reason why the ECB opted to
pursue a consensus model in its decision-making (Claeys and Linta, 2019). The idea was that if the ECB
was to speak for the euro area, as indeed dictated by its mandate, then it had to ignore national
preferences. And to be able to convince the public, it would have to speak with one voice. The decision
process and the communication of that decision would therefore not provide any information on
possible disagreements or how strong a consensus would lead to it. Also, even if the optimal size of a
committee is shown to be between 5-9, the ECB could not afford not to include all national central
banks in its decision body.
But the question that we then ask is whether the set-up of any given committee actually leads to very
different ways of deliberating. Ehrmann and Fratzscher (2007) show that there is not a single best
approach for central banks to adopt. Different ways of combining more or less individualistic
communication and decision-making strategies may deliver similarly good results in terms of financial
markets responsiveness and policy decisions’ predictability. Similarly, Riboni and Ruge-Murcia (2010)
look at five different Central Banks (BoC, BoE, ECB, Fed, SRC) and indicate that, despite having different
formal committee types, all central banks seem to follow a consensus model in the way they take
actual interest rate decisions.
We take a closer look at the way the BoE and the ECB reach their decisions and actually confirm that,
irrespective of the set-up, they both show a very similar trend in consensus building when they take
decisions.
Figure 1a and 1b plot all rate decisions since the late 90s for both the Bank of England and the ECB.
Also, we report the degree of agreement reached in each decision. Data is available for the Bank of
England as votes published after each meeting. This, however, is not available for the ECB, since it
does not publish votes. Hence, we rely on the methodology by Claeys and Linta (2019) who gather
information from various sources on how decisions were taken.2 We observe that in the current
juncture the BoE faces greater disagreements than in earlier times.
2
These include the transcripts of the press conferences following the governing council ‘monetary policy’ meetings which
also include transcripts of the Q&A with journalists, and, since 2015, the ‘accounts’, i.e. the summaries of the discussion of
the monetary policy meetings published by the ECB. Exceptional press releases may also be considered.
4Figure 1a: MPC voting decisions on the bank rate Figure 1b: ECB Governing Council voting decisions on
the interest rates - main refinancing rate (MRO) and
deposit facility rate (DFR)
Bank rate unanimity
strong majority 8/1 strong majority % MRO/DFR unanimous consensus majority
majority Introduction of
8% 6 negative interest
7
ra tes via DFR
7% 5 6
6%
4 5
5%
3 4
4%
2 3
3%
1 2
2%
0 1
1%
0% -1 0
Source: Bruegel based on Bank of England. Source: Bruegel based on ECB.
Notes: period from June 1997 to Feb 2022. Unanimity = all votes in support, Notes: period from Dec 1998 to Feb 2022. Before 5 June 2014, rate plotted
Strong majority 8/1 = all - 1 one vote, Strong majority = between strong corresponds to the MRO, after that to the DFR. The classification of the
majority 8/1 and majority, Majority = half + 1 votes in support. decision was made according to what was indicated by the ECB president in
the press conferences and by the monetary policy accounts, based on Claeys
and Linta (2019).
For more than half (61%) of the ECB meetings we have no information on the degree of consensus
reached. Figure 2 presents the summary of the degree of disagreement in both central banks
deliberations. Despite the ECB decision-making process relying on consensus, we observe that the
degree of unanimity reached in the two banks is quite similar when it comes to interest rate decisions.
When it comes to the unconventional measures, the BoE members reach consensus much faster. That
is to be expected as when it comes to bond purchases the ECB buys a much wider portfolio of bonds
with different risks.
Figure 2a: BoE MPC bank rate and bond purchase decisions and direction of the policy
Type of decision on interest rates Type of decision on bond purchases
5% 1%
8%
22%
13%
52%
21%
78%
unanimity strong majority 8/1 strong majority majority unanimity strong majority 8/1 strong majority majority
Source: Bruegel based on Bank of England. Notes: period from June 1997 to Feb 2022. Unanimity = all votes in support, Strong majority 8/1 = all - 1
one vote, Strong majority = between strong majority 8/1 and majority, Majority = half + 1 votes in support.
Figure 2b: ECB Governing Council interest rates and other decisions and direction of the policy
5Type of decision on interest rates Type of decision on other policies
7%
26%
38%
37%
56%
36%
Unanimity Consensus Majority Unanimity Consensus Majority
Source: Bruegel based on ECB and Claeys and Linta (2019). Notes: period from Dec 1998 to Feb 2022. The classification of the decision was made
according to what was indicated by the ECB president in the press conferences and by the monetary policy accounts. Note that, for the BoE chart we
show decisions on bond purchases, while for the ECB “other” includes more policies, e.g. forward guidance or credit operations.
The one characteristic the two Central Banks share is that they reach tightening interest rate decisions
more unanimously than easing decisions as Figure 3 shows. We can also see that, when it comes to
maintaining the policy unchanged, the decision is backed by mostly all members.
Figure 3a: Bank of England – Direction of policy move
Easing Tightening Maintain
3%
5%
9%
31% 21%
28%
17%
43%
Rate 55%
decisions 18%
38% 30%
unanimity strong majority 8/1 strong majority majority unanimity strong majority 8/1 strong majority majority
unanimity strong majority 8/1 strong majority majority
0% 1%
7%
16%
45% 45%
Other
decisions
76%
9%
unanimity strong majority 8/1 strong majority majority unanimity strong majority 8/1 strong majority majority
Figure 3b: ECB – Direction of policy move
6Easing Tightening Maintain
7% 5%
30% 30%
34%
Rate 36%
57%
decisions 61%
40%
Unanimity Consensus Majority Unanimity Consensus Majority Unanimity Consensus Majority
9%
30%
38%
46%
Other
decisions
45%
32%
Unanimity Consensus Majority Unanimity Consensus Majority
Source: Bruegel based on ECB and Bank of England. Notes: period from June 1997 to Feb 2022 for BoE, from Dec 1998 to Feb 2022 for ECB. For the
ECB, the classification of the decision was made according to what was indicated by the ECB president in the press conferences and by the monetary
policy accounts.
Last, Table 2 summarises the degree of disagreement together with how successful the banks are at
the time of the meeting. As the Bank of England reveals not only the target but also the tolerance
bands, it is easy to define success as the inflation rate being inside those bands (inflation between 1
and 3 per cent). For the past 25 years3, inflation in the UK has been within those rates 67% of the time.
Since the ECB does not have such a clear definition, we present three alternative definitions. All three
of them are much tighter than the one that the BoE uses and success is also less frequent.
We make the following general observations:
• The most relevant result is that the degree of Unanimity vs Majority in decisions is very similar
despite the very different decision-making set-up. Admittedly, we have a lot less observations
for the ECB and in any case, our numbers are at best a proxy rather than a measurement given
the absence of votes.
• What we label as consensus has a rather different meaning for the two Central Banks. For the
Bank of England, it is clearly a measurement of the strength of majority. But for the ECB it is
truly consensus in that it represents where the members find and agreement that is
acceptable to all, and not necessarily a vote between two options. In this respect they are not
directly comparable.
• Also, neither of the banks is affected in the way they reach a unanimous decision, by whether
they observe at time t, a successful inflation rate (i.e. conditional probability of unanimity
given that they observe S or NS). Observing a successful inflation rate, members come into
the meeting with a 50-50 chance of reaching unanimity.
3
We are aware that there was a change in the target in 2003, but given that it was a technical adjustment, we
considered the current indicator (Consumer Prices Index) and the 2% target and extended it backwards in our
calculations.
7Table 2: Bank of England and ECB - voting patterns and success
ECB BoE
Success Success Success
classification 1 classification 2 classification 3
S 17% 24% 36% 67%
Considering all
B 40% 34% 22% 17%
Governing Council
meetings A 42% 42% 42% 16%
NS 83% 76% 64% 33%
U 56% 56% 56% 53%
SM 8/1 - - - 21%
SM 6-7 - - - 22%
SM / C 37% 37% 37% 43%
Considering only
M 7% 7% 7% 5%
known decisions
P(U&S)/P(S) 53% 61% 50% 53%
P(U&NS)/P(NS) 57% 55% 59% 52%
P(M&S)/P(S) 0% 0% 3% 5%
P(M&NS)/P(NS) 8% 9% 9% 5%
Sources: Bruegel based on ECB, Eurostat and Bank of England.
Notes: S = Success in keeping inflation within the target band, B = Below the target band, A = Above the target
band, NS = A+B.
For ECB (Dec 1998 to Feb 2022): There was a change in the target due to the revision of the ECB strategy
announced on 8 July 2022 and effective as of the GovC meeting of 22 July 2022 - change in range presented in
(brackets). Success classification 1 considers ‘success’ if the observed inflation rate is between 1.8 and 2 (2.2),
Success classification 2 if between 1.6 and 2 (2.4), Success classification 3 if between 1 and 2 (3). U = Unanimity,
C = Consensus, M = Majority. When considering all decisions: 289 observations, when considering only known
ones: 112 observations.
For BoE (June 1997 to Feb 2022): It is considered ‘success’ when the observed inflation rate is between 1 and 3,
as officially set by the BoE. U = Unanimity (9-0 votes), SM 8/1 = Strong Majority (8-1 votes), SM 6-7 = Strong
Majority (7-2 or 6-3 votes), SM = SM 8/1 + SM 6-7, M = Majority (5-4 votes). Nr observations: 259.
Conclusions
Decision-making in Central Banks is done by committees as the institutions became independent from
their governments. Beyond that, there are many differences in the way the committees are set-up
and decisions are taken. Observing the ECB and the Bank of England, we conclude however, that they
value equally reaching decisions by unanimity. This is perhaps not surprising given the impact that
Central Bank actions have on the economy. Central Banks want to provide a clear signal in order to be
convincing.
References
Badinger, Harald, and Volker Nitsch. 2011. ‘National Representation in Multinational Institutions:
The Case of the European Central Bank’. Working Paper No. 3573. CESifo Working Papers.
https://www.ssrn.com/abstract=1927820.
8Bennani, H., T. Kranz, and M. Neuenkirch. 2018. ‘Disagreement between FOMC Members and the
Fed’s Staff: New Insights Based on a Counterfactual Interest Rate’. Journal of
Macroeconomics 58: 139–53. https://doi.org/10.1016/j.jmacro.2018.09.002.
Berger, Helge, Volker Nitsch, and Tonny Lybek. 2008. ‘Central Bank Boards around the World: Why
Does Membership Size Differ?’ European Journal of Political Economy, Does central bank
independence still matter?, 24 (4): 817–32. https://doi.org/10.1016/j.ejpoleco.2008.04.003.
Blinder, Alan S. 2007. ‘Monetary Policy by Committee: Why and How?’ European Journal of Political
Economy, Central bank transparency and central bank communication, 23 (1): 106–23.
https://doi.org/10.1016/j.ejpoleco.2006.01.003.
Claeys, Gregory, and Tanja Linta. 2019. ‘The Evolution of the ECB Governing Council’s Decision-
Making | Bruegel’. June 2019. https://www.bruegel.org/2019/06/the-evolution-of-the-ecb-
governing-councils-decision-making/.
Downward, Paul, and Andrew Mearman. 2007. ‘Decision-Making at the Bank of England: A Critical
Appraisal’. Oxford Economic Papers 60 (October): 385–409.
https://doi.org/10.1093/oep/gpn009.
Ehrmann, Michael, and Marcel Fratzscher. 2007. ‘Communication by Central Bank Committee
Members: Different Strategies, Same Effectiveness?’ Journal of Money, Credit and Banking
39 (2–3): 509–41. https://doi.org/10.1111/j.0022-2879.2007.00034.x.
Eichler, S., T. Lähner, and F. Noth. 2018. ‘Regional Banking Instability and FOMC Voting’. Journal of
Banking and Finance 87: 282–92. https://doi.org/10.1016/j.jbankfin.2017.10.011.
Gerlach-Kristen, Petra. 2009. ‘Outsiders at the Bank of England’s MPC’. Journal of Money, Credit and
Banking 41 (6): 1099–1115. https://doi.org/10.1111/j.1538-4616.2009.00248.x.
Hansen, Stephen, Michael McMahon, and Carlos Velasco Rivera. 2014. ‘Preferences or Private
Assessments on a Monetary Policy Committee?’ Journal of Monetary Economics 67: 16–32.
https://doi.org/10.1016/j.jmoneco.2014.06.004.
Harris, Mark N., Paul Levine, and Christopher Spencer. 2011. ‘A Decade of Dissent: Explaining the
Dissent Voting Behavior of Bank of England MPC Members’. Public Choice 146 (3): 413–42.
https://doi.org/10.1007/s11127-010-9597-6.
Harris, Mark N., and Christopher Spencer. 2009. ‘The Policy Choices and Reaction Functions of Bank
of England MPC Members’. Southern Economic Journal 76 (2): 482–99.
https://doi.org/10.4284/sej.2009.76.2.482.
Maier, Philipp. 2010. ‘How Central Banks Take Decisions: An Analysis of Monetary Policy Meetings’.
In Challenges in Central Banking, edited by Pierre L. Siklos, Martin T. Bohl, and Mark E.
Wohar, 320–56. Cambridge: Cambridge University Press.
https://doi.org/10.1017/CBO9780511762802.011.
Malmendier, U., S. Nagel, and Z. Yan. 2021. ‘The Making of Hawks and Doves’. Journal of Monetary
Economics 117: 19–42. https://doi.org/10.1016/j.jmoneco.2020.04.002.
Riboni, Alessandro, and Francisco J Ruge-Murcia. 2010. ‘Monetary Policy by Committee: Consensus,
Chairman Dominance or Simple Majority?’ The Quarterly Journal of Economics 125 (1): 363–
416.
9Rieder, Kilian. 2021. ‘Monetary Policy Decision-Making by Committee: Why, When and How It Can
Work’. European Journal of Political Economy, August, 102091.
https://doi.org/10.1016/j.ejpoleco.2021.102091.
Smales, Lee A., and Nick Apergis. 2016. ‘The Influence of FOMC Member Characteristics on the
Monetary Policy Decision-Making Process’. Journal of Banking & Finance 64 (March): 216–
31. https://doi.org/10.1016/j.jbankfin.2015.12.002.
Tucker, Paul. 2008. ‘Remarks on “Making Monetary Policy by Committee”’. Bank of England
Quarterly Bulletin 48 (No. 3): 347–50.
10You can also read
